The Media Audit is a multimedia survey conducted for the purpose of developing statistically reliable information about the audience levels and audience characteristics of radio stations, local TV news programs, cable TV viewing, daily newspapers, weekly and monthly publications, the Internet, social media, mobile Internet usage, local media websites, direct mail and out-of-home media. The Media Audit also collects socioeconomic information, product buying plans and purchasing activity for numerous products and services, retail stores and financial institutions that can be used to define the quality of individual media audiences. In addition, the information can be used in defining customer profiles and consumer market shares for the many products, services, retail establishments and banking institutions that are covered in the survey.
All interviews are conducted by telephone five days a week, Tuesday through Saturday. Tuesday through Friday, data is collected during the day and evening. On Saturday, interviewing is collected during the day.
A new sample frame is initiated at the beginning of each new survey period. All working blocks, which apply to the survey geography, are included in the new sample frame. This insures that the entire survey area will be represented via the completed distribution of the sample to be surveyed. In addition, the telephone numbers are selected in proportion to the geographic distribution of the households on a county-by-county basis. In order to reduce calling order bias, the ending sample is arranged into replicates. Each replicate is a sub-sample of its own and is representative of the survey area as a whole. In addition, the telephone numbers within the replicate are randomly shuffled.
Some markets may be identified for a buffer sample in order to improve sample representation.
In order to categorize pre-determined open ended questions, The Media Audit utilizes a standard set of coding rules.
On completion of all interviews, the survey sample is weighted by age, gender, ethnicity, county, and household size 18+ population estimates to the up-dated U.S. Census data as provided by the Nielsen Company.
Interviewers are instructed to read each question on the questionnaire exactly as worded. If the respondent does not understand the question, the interviewers are instructed to reread the question and not to explain.
Approximately ten percent of all interviews are either monitored or validated by a return call to the person that was interviewed.
For a media (other than radio), business, or product to be included in the stub row of a report, 1% of the respondents (after weighting) must mention the media, business or product. The only exceptions to this are stub rows that are standard to all markets, or if a business and/or product is included in the second report of the year as a result of being included in the first report of the year. This is done for report consistency.
For radio stations, which appear above the line in the report (home market), a .5% or higher reach of the market is required in Section 1, Radio-Past 7 Day Cume and a .1% or higher in Section 2 , Radio Listen to Most Often. For radio stations which appear below the line in the report, a 1% or higher is required in Section 1 for Radio-Past 7 Day Cume and a .1% or higher in Section 2, Radio-Listen to Most Often.
In order for a media or business to be included in a non-standard banner, there must be a minimum of 25 in tab. However, The Media Audit is limited to 15 banner points per page. As a result, if the number of media and/or businesses exceed the available number of banner points, only the top performing media and/or businesses will be included.
Limitations of Data
This report is based on a random sample of all telephone households in the designated survey area. Non-telephone households are not included in the sample. These non-telephone households may have different preferences, habits and characteristics from those with telephones. All statistics are based on recall of the respondents interviewed. Only adult males and females that are 18 years of age or older who had the most recent birthday are interviewed. In addition, all households with a household member who is affiliated with a media are not included in the survey. All persons selected may not choose to complete an interview. Although errors may occur in the interview, in the recording of data and/or in the processing of information, controls are established and efforts are made to eliminate such errors.
Single Phase Methodology
The Media Audit uses a single-phase telephone data collection process, i.e. all of the data is collected in a single telephone interview. This assures the users of the data integrity. A two-phase or three-phase methodology where self-completed questionnaires are mailed out to the respondent following the completion of the telephone interview runs the risk of partial returns of questionnaires and/or diaries. To compensate for the non returned documents, data must be ascribed for close to half or more than half of the telephone interview base in order to have a like sample size for the entire survey.
The Media Audit only ascribes a response when a respondent doesn't know or refuses to answer questions relating to the following:
The non-response rate to all of these queries except Annual Household Income, are usually 5% or less. Non-response to the income question can be as much as 20%. The lower the ascription rate, the more reliable the information.
The Media Audit Questionnaire is standardized across all markets. The only things that change from one market to another are the names of the local stores, shopping centers and financial institutions. This allows users of the data to aggregate multiple markets together for regional analyses.
True Data Analyses
The Media Audit does not ascribe its data in any way to make its media ratings conform to another study's media ratings. The Media Audit ratings are tabulated as reported by the respondents interviewed in The Media Audit survey.
Depending on the market size, The Media Audit data is gathered within four to 12 week time frames and delivered within 30-45 days following the close of the interviewing process. This removes the problems caused by aggregating data that is gathered on a continual basis over a six-month time frame or even a twelve-month time period if you want to get the "benefit" of the full sample base. Discreet time specific measurements also enable you to see the changes that occur within local media format and program changes as well as retail shopping changes.
The Media Audit defines its response rates, ascription amounts, survey dates, ethnic make-up, respondent age and gender distribution and other methodological criteria in the first three pages of each report. In addition, The Media Audit discloses the respondent count for each demographic and qualitative target audience on each page of the report and in its software. No other media service goes to this depth of respondent count disclosure.
Media Neutral Data Base
Distribution of The Media Audit sample to the metro area does not favor broadcast media with differing signal / distribution strengths. This reduces the problem of applying indexing to non-concentric survey area data.
Local Media Web Site
The Media Audit collects web site ratings data for local newspapers, television stations, radio stations, city guides and alternative newsweeklies in each of the 80+ markets surveyed. The primary usage of this information is mostly relevant to local advertisers and advertising agencies. However, it is possible that local media web sites will become a local spot advertising medium for national advertisers and their agencies as the medium grows in importance and acceptance.